Wiktor Rafał Żelazny

The remote sensing of the biophysical and biochemical parameters of crops facilitates the preparation of application maps for variable-rate nitrogen fertilization. According to comparative studies of machine learning algorithms, Gaussian process regression (GPR) can outperform more popular methods in the prediction of crop status from hyperspectral data. The present study evaluates GPR model accuracy in the context of spring wheat dry matter, nitrogen content, and nitrogen uptake estimation. Models with the squared exponential covariance function were trained on images from two hyperspectral cameras (a frenchFabry–Pérot interferometer camera and a push-broom scanner). The most accurate predictions were obtained for nitrogen uptake (${\mathrm{R}}^2=0.75–0.85$, ${\mathrm{RPD}}_{\mathrm{P}}=2.0–2.6$). Modifications of the basic workflow were then evaluated: the removal of soil pixels from the images prior to the training, data fusion with apparent soil electrical conductivity measurements, and replacing the Euclidean distance in the GPR covariance function with the spectral angle distance. Of these, the data fusion improved the performance while predicting nitrogen uptake and nitrogen content. The estimation accuracy of the latter parameter varied considerably across the two hyperspectral cameras. Satisfactory nitrogen content predictions (${\mathrm{R}}^2>0.8$, ${\mathrm{RPD}}_{\mathrm{P}}>2.4$) were obtained only in the data-fusion scenario, and only with a high spectral resolution push-broom device capable of capturing longer wavelengths, up to 1000 $\mathrm{n}$$\mathrm{m}$, while the full-frame camera spectral limit was 790 $\mathrm{n}$$\mathrm{m}$. The prediction performance and uncertainty metrics indicated the suitability of the models for precision agriculture applications. Moreover, the spatial patterns that emerged in the generated crop parameter maps accurately reflected the fertilization levels applied across the experimental area as well as the background variation of the abiotic growth conditions, further corroborating this conclusion.

There is a need to minimize the usage of traditional laboratory reference methods in favor of spectroscopy for routine soil carbon monitoring, with potential cost savings existing especially for labile pools. Mid-infrared spectroscopy has been associated with accurate soil carbon predictions, but the method has not been researched extensively in connection to C lability. More studies are also needed on reducing the numbers of samples and on how to account for the compositional nature of C pools. This study compares performance of two classes of partial least squares regression models to predict soil carbon in a global (models trained to data from a spectral library), local (models trained to data from a target area), and calibration-spiking (spectral library augmented with target-area spectra) scheme. Topsoil samples were+ scanned with a Fourier-transform infrared spectrometer, total and hot-water extractable carbon determined, and isometric log-ratio coordinates derived from the latter measurements. The best RMSEP was estimated as 0.38 and 0.23 percentage points $\mathrm{TC}$ for the district and field scale, respectively—values sufficiently low to make only qualitative predictions according to the RPD and RPIQ criteria. Models estimating soil carbon lability performed unsatisfactorily, presumably due to low labile pool concentration. Traditional weighing of spiking samples by including multiple copies thereof in training data yielded better results than canonical partial least squares regression modeling with embedded weighing. Although local modeling was associated with the most accurate predictions, calibration spiking addressed better the trade-off between data acquisition costs and model quality. Calibration spiking with compositional data analysis is, therefore, recommended for routine monitoring.

Hyperspectral imaging (HSI) has been gaining recognition as a promising proximal and remote sensing technique for crop drought stress detection. A modelling approach accounting for the treatment effects on the stress indicators’ standard deviations was applied to proximal images of oilseed rape—a crop subjected to various HSI studies, with the exception of drought. The aim of the present study was to determine the spectral responses of two cultivars, ‘Cadeli’ and ‘Viking’, representing distinctive water management strategies, to three types of watering regimes. Hyperspectral data cubes were acquired at the leaf level using a 2D frame camera. The influence of the experimental factors on the extent of leaf discolorations, vegetation index values, and principal component scores was investigated using Bayesian linear models. Clear treatment effects were obtained primarily for the vegetation indexes with respect to the watering regimes. The mean values of RGI, MTCI, RNDVI, and GI responded to the difference between the well-watered and water-deprived plants. The RGI index excelled among them in terms of effect strengths, which amounted to $-0.96[-2.21,0.21]$ and $-0.71[-1.97,0.49]$ units for each cultivar. A consistent increase in the multiple index standard deviations, especially RGI, PSRI, TCARI, and TCARI/OSAVI, was associated with worsening of the hydric regime. These increases were captured not only for the dry treatment but also for the plants subjected to regeneration after a drought episode, particularly by PSRI (a multiplicative effect of $0.33[0.16,0.68]$ for ‘Cadeli’). This result suggests a higher sensitivity of the vegetation index variability measures relative to the means in the context of the oilseed rape drought stress diagnosis and justifies the application of HSI to capture these effects. RGI is an index deserving additional scrutiny in future studies, as both its mean and standard deviation were affected by the watering regimes.